Methods of distinguishing biologically active cannabis and fibre cannabis


Samples examined
Extraction and fractionation
Gas-chromatographical analyses: findings and discussion


Pages: 55 to 59
Creation Date: 1968/01/01

Methods of distinguishing biologically active cannabis and fibre cannabis

Francesco TOFFOLI
Istituto Superiore di Sanité, Rome


By applying gas chromatography at programmed temperatures ranging from 106° to 268 °C (fig. 1) to the respective petrol-ether extracts, it is possible to distinguish cannabis indica (including that grown in Italy) and the resin of cannabis indica (hashish) from cannabis grown for fibre (fig. 3). The method used consisted in analysing the chromatogram fraction comprised between 230° and 245 °C, containing the peaks relating to the characteristic components (acid and phenolic) of the drug.

So far, botany and pharmacognosy do not provide sufficient elements to evaluate drugs obtained from cannabis indica (or cannabis sativa var. indica) or even to distinguish those derived from the flower and leaf of cannabis grown for fibre. A reliable result may be obtained from a pharmacological comparison carried out preferably on dogs [ 1] - [ 3] , [ 5] - [ 8] , [ 11] , [ 12] , [ 15] , [ 17] , [ 26] ; but such experiments require a relatively large amount of the drug: a few grams of resin (hashish) or decagrams of flower or leaf.

Gas chromatography, on the other hand, can be applied to an extremely small sample, and in a remarkably short time yields an impressive amount of qualitative and quantitative information [ 4] , [ 13] , [ 14] , [ 25] .

Although research on the chemistry of the active principles of Indian hemp is far advanced, especially as regards position and steric isometry, it is not yet possible to attribute a definite pharmacological activity to each component, or, hence, to calculate the activity of the drug from its chemical composition. For these reasons, any results produced by a spectrophotometric or chromatographic analysis on paper, on thin layer or in a liquid gas column - should always be regarded with caution and from an empirical angle, since each spot or peak may be produced by one or other of the various isomers or by mixtures of these in an unknown proportion.

Note: The original of this article is in Italian.

In this preliminary investigation, 1 we have confined our attention to examining by gas chromatography, at a programmed and constant temperature, the following group of substances, all of them closely related: cannabidiols, cannabinols and tetrahydrocannabinols and relative carboxylated acid products, which include substances with a sedative and narcotic action.

The carboxylic acids are not extracted from the petrol-ether cannabis extract with sodium bicarbonate and carbonate solutions [ 23] . They are extracted only with the sodium hydroxide solutions which, on the other hand, do not extract the phenols- these remain in the petroleum ether together with the neutral substances [ 22] . 2 As regards the acids, this phenomenon has been attributed to a steric hindrance or to a chelation relation between the carboxyls and other oxygenized functions present in the molecule [ 23] : at any rate, in some cases the carboxylic acid, which is extractable only with sodium hydroxide, can be extracted from the bicarbonate solutions after the opening of the oxygenized hexagonal bond [ 23] ; in the case of the phenols, the phenomenon referred to has been attributed to a steric hindrance.

The authors of this survey, on the other hand, starting out from the hypothesis that such behaviour, whether on the part of the acids or the phenols, is simply a function of their low solubility in water, found that by subsequently treating the petrol-ether solution with sodium hydroxide solutions in 50% alcohol, it is also possible to extract the phenols and thus to separate these from the neutral substances.

1The results of subsequent investigations may be communicated later.

2For purposes of this study, we have attributed the conventional meaning to the terms "acids" and "phenols ", inasmuch as we have not shown that all, and only, the acids are extracted by the aqueous sodium hydroxide, or that as a result of subsequent treatment with hydroalcoholic sodium hydroxide, all, and only, the phenols are extracted.

More accurate methods of separating and examining the acid and phenolic fractions present in the drug are now in progress, and the results will be available later.

Samples examined

The material analysed comprised five samples of cannabis indica resin (hashish), supplied by the United Nations Division of Narcotic Drugs, Geneva, and labelled as follows:


and two hashish samples seized by the police:

Unknown origin

In addition, ten samples of cannabis indica, consisting of the flower and leaf of the female plant, were examined, the first three originating from plantations in Italy, the others being supplied by the United Nations Division of Narcotic Drugs, Geneva:

  1. Naples, 1964: Plants cultivated at the Experimental Station for Officinal Plants attached to the Botanical Institute of the University of Naples, obtained through successive generations since 1931, from seeds originating from Calcutta [ 3] , [ 16] , [ 24] ;

  2. Rome, University, 1964: Plants obtained in Rome from seeds obtained from the Naples Experimental Station for Officinal Plants;

  3. Rome, Higher Institute of Health: Plants obtained from seeds seized in the winter of 1963/1964 and originating from a plantation at Eboli (Salerno);

4. Morocco
UNC 21
5. Brazil
UNC 11
6. Brazil
UNC 13
7. Brazil
UNC 14
8. Brazil
UNC 15
9. Brazil
UNC 17
10. Brazil
UNC 19

One, London 8028, of unknown origin, had been seized by the police.

Eight varieties of cannabis grown for fibre were also examined, four of them:

6 1959
Experimental, Germany
Geneva Botanical Gardens
28 1960
Experimental, Germany

being supplied by the United Nations Division of Narcotic Drugs, Geneva, and four (Fibranova, Eletta Campana, Carmagnola, T 4) by the Caserta Office of the National Association of Hemp Growers. The four latter were harvested in 1964 during and after the flowering stage.

Extraction and fractionation

The air-dried drug is extracted 5-6 times in a mortar with sufficient volumes of petroleum ether, boiling point 30-50 °, to cover the drug itself; it is filtered through a glass paste under gentle suction. The extract is evaporated under vacuum until a concentration is obtained corresponding to 20 g of drug per 100 ml of solution.

On the basis of a fibre cannabis sample (Carmagnola) and one of cannabis indica grown in Italy, a petrolether extract was also prepared, by the same methods, one of the leaf tops and one of the flowering tops.

On the basis of the total extract (E), the substances soluble in aqueous 2 N NaOH were separated from those soluble in 4 % NaOH in 50 % alcohol. The results are shown in the following diagram:

Full size image: 2 kB Full size image: 8 kB

The total petrol-ether solution (E) is extracted twice with aqueous 2N NaOH, with appropriate cold washings; the aqueous fraction, acidified with HCI (still cold), is extracted twice with petroleum-ether. The petrol-ether phase thus obtained, containing the acid fractions of the resin, is dehydrated with anhydrous Na 2SO 4, filtered and concentrated under vacuum to the same volume of the total E extract (E 2).

The E 1 petrol-ether solution representing the residue from the foregoing extraction process, is extracted three times with 4% NaOH and 50% EtOH, using the same methods as described above. The neutral components of the drug remain in the petroleum-ether (E 1-E 3). After acidification, the hydro-alcoholic phases are re-extracted with petroleum-ether. This petrol-ether fraction, predominantly containing substances of a phenolic nature, is dehydrated with anhydrous Na 2SO 4, filtered and condensed under vacuum (E 3) to the volume of the initial extract.

The E 2 (acid) and E 3 (phenolic) fractions were subjected to the colour reaction test with ethyl alcohol saturated with HC1, which Nickolls has described for the acid components only (a) (19, 21), and which we have extended to the phenolic fraction (ph).

The reaction was invariably positive for both fractions (a and ph) for the resins and female flowers of cannabis indica, including those grown in Italy, and invariably negative for cannabis grown for fibre.

The intensity of the colour obtained depends in general on the quantity of active principles contained in the drug. The tint is related to the species of drug: for the acid fraction, it is pink for cannabis indica, and pinkish-violet for the resins of cannabis indica; for the phenolic fraction, the colours are less pure and less constant.

Our findings fully bear out the results obtained by I. Novak (20) with the Nickolls reagent for Hungarian fibre cannabis and hashish.

A single sample of cannabis indica (London 8028), of unknown origin, although obviously exerting a biological action on dogs, did not produce such a colour reaction.

Gas-chromatographical analyses: findings and discussion

The gas chromatograph used was an R.S. Co.mod. 1601/2, with a detector and flame ionization.

The identification test was carried out in each case with 2µl of petrol-ether solution (100 ml of which correspond to 20 g of drug), using a column 1.8 m high and 3 mm in diameter, containing SE 30 (3 %) on Chromosorb W (80-100 mesh). The carrier gas used was nitrogen with a rate of flow of 25 ml/min. Initially, the temperature was maintained constant, for 5' at 106 °C; then increased to 230 °C at a rate of 10 °/min, and then to 268 °C, at 2.5 °/min, in order to obtain a better separation of the fraction containing the cannabinolic derivatives of the drug.

A diagram of a chromatogram relating to a cannabis indica resin is shown in figure 1.

Fig. 1. Gas chromatogram of the petrol-ether extract (E) of a sample of resin obtained from Cannabis indica (Greece UNC 1A)

The fraction comprised between 230° and 245° contains the characteristic components of the drug, composed of acids and phenols. It contains no neutral substance. The low-boiling (230°) and high-boiling (>245°) fractions, on the other hand, do not contain substances of a phenolic and acid nature but only neutral components.

In the aforementioned 230°-245° fraction, the samples examined (resin of cannabis indica, cannabis indica, fibre cannabis) give four main peaks composed, inter alia, according to Lerner (14), of cannabidiol, tetrahydrocannabinol and cannabinol.

Peak II, which can probably be identified with the cannabichromene isolated by Mechoulam (information supplied by the author), is always relatively small and not even present in all the samples; it will not be considered here, as it is not useful in identifying the drug.

Peak III probably corresponds to r l-2 - tetrahydrocannabinol; it differs from the rl-6 - tetrahydrocannabinol supplied to us by Mechoulam, and which we have used as a standard, using the same columns as already described, at a constant temperature of 212 °C, with a nitrogen flow of 30 ml/min (fig. 2). Peaks I and IV are probably partly attributable to cannabidiol and cannabinol, as well as to the corresponding acids.

The acid fraction gives the same peaks as the phenolic fraction: evidently the acids undergo decarboxylation during gas chromatography (4,23), to give the same three or four phenols contained in the phenolic fraction and pre-existing in the drug (or possibly formed as a result of decarboxylation by the alkalis.

The ratios between the areas of the three peaks obtained by working at a programmed temperature are characteristic for some groups of samples. This provides a guide to the classification of the cannabis examined (resin of cannabis indica, flower and leaf of cannabis indica, flower and leaf of fibre cannabis) as may be deduced from figure 3. In the latter, the ratios between the areas of the three components in the 230°-245° fraction, taking their sum as equal to unity, are given in a ternary diagram and indicated by I, III and IV, on the basis of the increasing retention times. No account is taken of the absolute quantity of the active principles.

Fig. 2. Gas chromatogram of the acid and phenolic fraction (230°-245°) (unbroken line) and of the (1.6)- tetrahydrocannabinol (hatched line)

Fig. 3. Sweden UNC 7; Greece UNC 18; Greece UNC lC; Origin unknown 6786; Greece UNC 1A; Libya 7656; Greece UNC 1D

Morocco UNC 21; Brazil UNC 11; Brazil UNC 13; Brazil UNC 14;

Brazil UNC 15; Brazil UNC 17 and Brazil UNC 19. Naples University, Rome University and the Higher Institute of Health, Rome

Germany UNC 23: Caserta Fibranova ; Germany UNC 27: Caserta T 4; Cyprus UNC 6: Caserta E.C. and Geneva UNC 28: Caserta Carmagnola

Composition of the petrol-ether extract of Cannabis indica and fibre cannabis. The extracts of Cannabis indica and of the resin of Cannabis indica invariably contain high proportions of components III and IV; the fibre cannabis extract is composed almost exclusively of the I component

Full size image: 15 kB

As will be noted, all the species of fibre cannabis contain almost exclusively substances of an acid nature which give rise to peak I. As regards cannabis indica and the resins of cannabis indica, these are present either in a negligible amount, or together with substances that give rise to peaks III and IV, the latter including those responsible for the pharmacological activity of the drug.

In the case of a drug of unknown origin (London 8028), for which no diagram has been included, the ratios between the areas of the three main components (I, III, IV) approximate to those of the cannabis indica grown in Italy (3:10:1).

As regards the total of fibre cannabis extract (E), no differences emerged even at the various ripening stages (unripe, ripe and overripe) between the ratios of the components of the 230°-245° fraction, but only a slight enrichment of the high-boiling (>245°) fraction with the progress of ripening.

Full size image: 14 kB

On the basis of a sample of fibre cannabis (Carmagnola) and one of cannabis indica (Rome University), extracts of the leaf tops and flowering tops were chromatographed separately. As regards the former, the ratio between the components of the 230°-245° fraction was the same for the leaves and the flowers; but in the flowers the components were present in a proportion seven times higher than for the leaves [ 9] . In the case of cannabis indica, on the other hand, the flowering tops mainly contain components I, and the leaves components III. These, as already stated, presumably contain one of the substances mainly responsible for the biological activity of the drug. This is confirmed by the fact that, other conditions being equal, the leaf extract produces a much more clearly marked ataxia in the dog (U.S.P. X) than that caused by the flowers; consequently, not all the biological activity of the plant can be ascribed to the latter [ 9] , [ 11] .


By comparing the results of applying gas chromatography to:

  1. The total petrol-ether extract;

  2. The fraction extracted from sodium hydroxide and

  3. The fraction extracted from sodium hydroxide dissolved in approximately 50% alcohol;

we obtained the following results:

  1. The fraction containing the cannabinolic derivatives exhibits four peaks in gas chromatography, of which one-II-is invariably of very modest proportions compared with the others (I, III, IV).

  2. By means of extractions ( b) and ( c) respectively, it was found that the three peaks of the total extract corre spond exactly to the three peaks of the acid fraction (a) and the three peaks of the phenolic fraction; hence the presence is deduced, in the extracts, of at least three acids, Ia, IIIa and IVa, which in gas chromatography are decarboxylized to give at least three phenols, Iph, IIIph and IVph, which are identical with those preexisting in the extract.

Comparison with a tetrahydrocannabinol revealed that III is probably a tetrahydrocannabinol and IIIa a tetrahydrocannabinolic acid.

  1. The diagrams contained in figure 2 show the predominance, in the case of fibre cannabis, of peak I which, under the experimental conditions adopted by use, is almost entirely of an "acid" nature; cannabis indica and hashish invariably exhibit - together with variable proportions of peak I-peaks III and IV (either of acid or phenolic origin), relating to the components considered to be pharmacologically active.

Our thanks are due to Dr. R. Mechoulam of the Weizmann Institute, Rehovoth, Israel, for supplying us with products and information.

We are also indebted to Dr. G. Liguori and his colleagues Drs. Mellisurgo and Tedeschi of the Caserta Office of the National Association of Hemp Growers, for the supply of hemp samples.

Rome, Istituto Superiore di Sanita (Higher Institute of Health), Biology Laboratories, 1966.



Balozet, L., Arch. Inst. Pasteur, Tunis, 26 318, 1937.


Bose, B. C., et al, Arch. int. Pharm. Therapie 146 99, 1963; ibid., 147 285, 291, 1964.


Carbonaro, G., and Imbesi A., Boll. Soc. It. Biol. Sper. 17 406, 1942.


Davis, T. W. M., Farmilo C. G., and Osadchuk M ., Anal. Chem. 35 751, 1963.


De Ropp, R. S., Journ. Am. Pharm. Ass., Sci. Ed. 49 756, 1960.


Do Valle, R. J., Rev. Inst. Adolfo Lutz 21 83, 1961.


Gaoni, Y., and Mechoulam, R., Journ. Am. Chem. Soc. 86 1646, 1964.


Grlic, L., Bull. Narc. 14 (3) 37, 1962.


Grlic, L., Bull. Narc. 16 (4) 29, 1964.


Hively, R. L., Mosher, W. A., and Hoffmann, F. W., Journ. Am. Chem. Soc. 88 1832, 1966.


Imbesi, A., Atti Soc. Pelor. Sci. Fis. Mat. Nat. 4 373, 1957/1958.


Kechatov, E. A., Bull. Narc. 11 (4) 5, 1959.


Kingston, C. R. and Kirk, P. C., Anal. Chem. 33 1794, 1961.


Lerner, M., Science 140 175, 1963.


Loewe, S., Science 102 615, 1945.


Longo, B., Boll. Orto Bot. University of Naples 13 17, 1934.


Malowan, L. S., Ber. Pharm. Gesell. 48, 150, 1938.


Mechoulam, R., and Gaoni Y., Journ. Am. Chem. Soc. 87 3273, 1965.


Nickolls, L. C., Analyst 61 604, 1936.


Novak, I., Buzas G., and Toth, L., Pharmazie 17 166, 1962.


Rathenasinkam, E., Analyst 73 509, 1948.


Scaringelli, F., Journ. Ass. Off. Agric. Chemists 44 296, 1961.


Schultz, O. E., and Haffner, G., Arch. Pharm. 291/63 391, 1958.


Susanna, V., Boll. Soc. It. Biol. Sper. 23 668, 1948.


Taylor, E. C., Lenard K., and Shvo, Y., Journ. Am. Chem. Soc. 88 367, 1966.


Walton, R. P., Martin L. F., and Keller J. H., Journ. Pharm. Exper. Therap. 62 239, 1938.